My goal is to understand the neural mechanisms underlying contour extraction, the presumed first step in visual object recognition. The primary visual cortex (V1) has the necessary response properties, connectivity patterns and functional organization to support the extraction of a contour through nonlinear interactions amongst neurons responding to its component elements (short line segments). I propose to test whether V1 has specific tuning for contours by measuring evoked neural activity as I parametrically vary contour salience through manipulation of contour geometry. Specifically, I will test the hypothesis that the response nonlinearites in V1 closely parallel contour salience. I will use a combination of optical imaging and electrode recording in alert behaving macaques. This technique allows direct visualization and comparison of cortical responses to different stimuli and reveals their relationship to the underlying functional maps. The results of this study, understanding the extraction and processing of contours, are prerequisite for the development of visual prostheses for the blind and bear general relevance to understanding cortical disorders of vision.